CORPORATE SOCIAL RESPONSIBILITY AND SHAREHOLDER …sites.bu.edu/cflammer/files/2018/09/Flammer_AMJ2013.pdf · that external pressure to becoming green (e.g., en-vironmental regulations,

CORPORATE SOCIAL RESPONSIBILITY ANDSHAREHOLDER REACTION: THE

ENVIRONMENTAL AWARENESS OF INVESTORS

CAROLINE FLAMMERUniversity of Western Ontario

This study examines whether shareholders are sensitive to corporations’ environmen-tal footprint. Specifically, I conduct an event study around the announcement ofcorporate news related to environment for all US publicly traded companies from 1980to 2009. In keeping with the view that environmental corporate social responsibility(CSR) generates new and competitive resources for firms, I find that companiesreported to behave responsibly toward the environment experience a significant stockprice increase, whereas firms that behave irresponsibly face a significant decrease.Extending this view of “environment-as-a-resource,” I posit that the value of environ-mental CSR depends on external and internal moderators. First, I argue that externalpressure to behave responsibly towards the environment―which has increased dra-matically over recent decades―exacerbates the punishment for eco-harmful behaviorand reduces the reward for eco-friendly initiatives. This argument is supported by thedata: over time, the negative stock market reaction to eco-harmful behavior hasincreased, while the positive reaction to eco-friendly initiatives has decreased. Second,I argue that environmental CSR is a resource with decreasing marginal returns andinsurance-like features. In keeping with this view, I find that the positive (negative)stock market reaction to eco-friendly (-harmful) events is smaller for companies withhigher levels of environmental CSR.

Corporate social responsibility (CSR) has re-ceived increasing attention in the past decades,both among practitioners and in the academic lit-erature. While the original focus of CSR was on“social” responsibility (e.g., paying fair wages toemployees, community-based programs), a recentdevelopment is the inclusion of environmental re-sponsibility (e.g., the reduction of CO2 emissions).This “environmental CSR” is becoming an integralpart of CSR and plays an increasingly importantrole in the corporate landscape. For example, in a

recent survey of 766 CEOs conducted by Accentureand United Nations Global Compact (UNGC), 93percent of the CEOs surveyed believe that sustain-ability will be critical to the future success of theirbusinesses, and 91 percent report that their com-pany will employ new technologies (e.g., renew-able energy) to address sustainability issues overthe next five years (Accenture & UNGC, 2010).

The increasing importance of environmental CSRamong practitioners is receiving considerable at-tention in academic research. A growing literaturestudies the reasons why companies engage in envi-ronmental CSR and how it relates to corporate per-formance (e.g., for recent reviews, see Ambec andLanoie [2008], Berchicci and King [2007], and Et-zion [2007]). In particular, Hamilton (1995), Klas-sen and McLaughlin (1996), and Shane and Spicer(1983) examined the relationship between environ-mental and stock-market performance. While thesearticles point toward a positive relationship be-tween environmental CSR and stock prices, little isknown about whether and how this relationshiphas evolved over time.

Regarding the stock market’s reaction to environ-mental CSR, perhaps one of the most prominent

I am grateful to Associate Editor Yan (Anthea) Zhangand three anonymous reviewers whose comments andguidance substantially improved this article. I also thankseminar participants at MIT, IAE Business School, KeanUniversity, as well as participants at the Alliance forResearch on Corporate Sustainability Conference (2012),the IBFR Global Conference on Business and Finance(2011), and the Swiss Economists Abroad Conference(2011) for helpful comments and suggestions. Zeina Bar-ton, Vanessa Mortillaro, and Magdalena Oleksiewicz pro-vided excellent research assistance. Financial supportfrom Kean University’s Office of Research & SponsoredPrograms is gratefully acknowledged.

758

� Academy of Management Journal2013, Vol. 56, No. 3, 758–781.http://dx.doi.org/10.5465/amj.2011.0744

Copyright of the Academy of Management, all rights reserved. Contents may not be copied, emailed, posted to a listserv, or otherwise transmitted without the copyright holder’s expresswritten permission. Users may print, download, or email articles for individual use only.

examples is British Petroleum’s (BP) oil spill inci-dent in April 2010. This oil spill contaminated alarge area of marine environment along the Gulf ofMexico, and is currently the biggest off-shore oilspill in US history. On the day of the incident, BP’sstock price was $59.5. By the end of June 2010, thestock price had dropped to $28.9―about half of itspre-incident value. As this example illustrates, en-vironmental issues can have dramatic implicationsfor stock prices. Yet, another set of anecdotes sug-gests that this may not always have been the case,or at least not in such magnitude. For instance,Exxon’s oil spill in March 1989 was considered oneof the most damaging incidents to the environment.However, Exxon’s stock price decreased only mar-ginally, from $44.5 on the day of the incident to$41.75 in April, and quickly recovering to its pre-incident level by June.

Arguably, BP and Exxon are very different com-panies, and such differences may partly explain thediffering stock-market reactions (e.g., they may dif-fer in their ability to manage public relations crises,or in the strength of their environmental manage-ment). Also, BP’s oil spill was of a more severemagnitude, which may have triggered a relativelystronger stock-market reaction. Nevertheless, thesearguments are unlikely to account for such largedifferences. Rather, these two examples suggestthat shareholders’ perception of environment-re-lated corporate behavior may have shifted consid-erably over the years.

In this study, I extend existing theories to derivehypotheses on how the relationship between envi-ronmental CSR and stock prices may have evolvedover time. I then systematically investigate whethershareholders reward or penalize corporations fortheir behavior toward the environment and howsuch rewards and punishments have changed overthe past decades.

To identify events that reveal information aboutthe firms’ environmental CSR, I searched the WallStreet Journal (WSJ) for relevant press coverage onresponsible and irresponsible behavior toward theenvironment for the whole universe of US publiclytraded companies from 1980 to 2009. I then ana-lyzed how the stock market reacted to these eventsby conducting an event study around the dates ofthe WSJ articles. I performed the analysis sepa-rately for the announcement of eco-friendly corpo-rate initiatives (e.g., the introduction of a recyclingprogram), and the announcement of eco-harmfulcorporate behavior (e.g., the release of hazardouswaste into the environment).

My conceptual framework builds on the argu-ment that environmental CSR generates new andcompetitive resources for firms. This argument isexemplified in Porter (1991), in instrumental stake-holder theory (e.g., Jones, 1995), in the natural re-source-based view of the firm (e.g., Hart, 1995;Russo & Fouts, 1997), and in the recent literature onsustainability in business (e.g., Clelland, Dean, &Douglas, 2000; Rusinko, 2007; Russo & Harrison,2005). In keeping with this argument, I found thatthe stock market reacted positively to the an-nouncement of eco-friendly initiatives, and nega-tively to the announcement of eco-harmfulbehavior.

I then extended this framework by assuming thatthe value of “environment-as-a-resource” dependson both external norms of environmental CSR andinternal levels of environmental CSR.

First, from an external perspective, I assumedthat external pressure to becoming green (e.g., en-vironmental regulations, media attention to the en-vironment, customers’ sensitivity to environment-related issues) sets the institutional norm ofenvironmental CSR. The more that becoming greenis institutionalized as the norm, the greater thenegative effect of negative news on perceptions of afirm, because firms are punished for not followingthe norm. Similarly, the more that companies enactthe institutional norm of going green, the less reac-tive shareholders are to the announcement of eco-friendly initiatives. I provide several “stylized”facts suggesting that external pressure―and hencethe norm of becoming green―has increased tre-mendously over recent decades. In keeping withthe above arguments, I have found that the positivestock-market reaction to eco-friendly initiatives hasdecreased over time, while the negative reaction toeco-harmful behavior has become more negative.

Second, from an internal perspective, I argue thatenvironmental CSR is a resource with decreasingmarginal returns. The higher the “stock” of thisresource, the lower the additional value generatedby additional investments in environmental CSR,and hence the lower the shareholders’ reward foreco-friendly initiatives. Similarly, the higher thestock of environmental CSR, the lower the punish-ment for eco-harmful behavior: a larger stock ofenvironmental resources may act as “insurance,”mitigating shareholders’ negative reaction to theannouncement of eco-harmful events. In support ofthese arguments, I find that firms with strongerenvironmental performance―measured by firm-level indexes of environmental strengths and con-

2013 759Flammer

cerns from Kinder, Lydenberg, Domini Research &Analytics (KLD)―experience a smaller stock-priceincrease following the announcement of eco-friendly initiatives as well as a smaller decreasefollowing the announcement of eco-harmfulbehavior.

Overall, the findings of this study support theview of environment-as-a-resource and shed lighton how the value of this resource depends on ex-ternal and internal moderators. In the followingsections, I develop my theoretical arguments indetail, describe the methodology, present the em-pirical results, and conclude by discussing the im-plications and limitations of my findings.

THEORY AND HYPOTHESES

Environmental CSR and Shareholder Reaction

The link between environment and managementhas been an active area of research. The early liter-ature, in the spirit of Friedman’s (1962, 1970: 122)view that the “social responsibility of business is toincrease its profits,” saw CSR as a cost of doingbusiness. CSR would decrease profits and therebyviolate the contractual relationship with sharehold-ers. For instance, the introduction of a new recy-cling program would require the installation of newphysical capital, the training of employees, and soon, all of which would be costly to the firm.

This view has been challenged in subsequentresearch. Freeman’s (1984) stakeholder theory sug-gested that companies should consider the interestsof a broader group of stakeholders—everyone whocan substantially affect, or be affected by, the wel-fare of a company. Several extensions of stake-holder theory have been proposed (for a review, seeAgle, Donaldson, Freeman, Jensen, Mitchell, andWood [2008]). In particular, in Jones’s (1995) in-strumental stakeholder theory, CSR efforts wereseen as potentially instrumental in obtaining nec-essary resources or stakeholder support. For exam-ple, the introduction of a new recycling programmight improve the company’s reputation and ap-peal to new customers and other stakeholders whoare concerned about the environment. Along simi-lar lines, in the aforementioned survey by Accen-ture and UNGC, 72 percent of the CEOs cited“brand, trust, and reputation” (2010: 14) as one ofthe main factors driving them to take action onsustainability issues. In their meta-analysis of theliterature, Orlitzky, Schmidt, and Rynes (2003) fur-ther emphasized that reputation may be an impor-

tant mediating variable of the relationship betweencorporate social responsibility and financialperformance.

Related literature in management further chal-lenged Friedman’s view. In particular, Porter(1991) argued that profitability and pollution re-duction might not be mutually exclusive goals. Inhis view, pollution is a waste of resources (e.g.,energy, material). Accordingly, efforts to reducepollution (e.g., through improved products or pro-cesses) might not only reduce a company’s environ-mental footprint but also strengthen its competi-tiveness. In a related argument, Porter (1991) andPorter and van der Linde (1995a, 1995b) proposedthat properly designed environmental regulationscan stimulate innovation and enhance competitive-ness. This proposition, known as the “Porter hy-pothesis,” has spurred a large debate in the litera-ture on environmental regulations (for a recentreview, see Ambec, Cohen, Elgie, and Lanoie[2011]).

A growing literature has extended Porter’s view(for examples of detailed reviews of this literature,see Ambec and Lanoie [2008], Berchicci and King[2007], and Etzion [2007]). For instance, the litera-ture on sustainability in business has examinedways in which companies can become more sus-tainable (i.e., “green”), and how these greening ini-tiatives influence financial performance. In partic-ular, it reports that companies can become moresustainable by leveraging, for example, the lowhanging fruits of efficiency and waste managementto achieve significant financial benefits (e.g., Clel-land et al., 2000; Rusinko, 2007; Russo & Harrison,2005). More complex initiatives studied includeefforts to integrate sustainability into product de-sign (e.g., Lenox, King, & Ehrenfeld, 2000; Waage,2007), to pursue environmental management sys-tems (e.g., Melnyk, Sroufe, & Calantone, 2003;Sroufe, 2003), and to “green” the supply chain (e.g.,Linton, Klassen, & Jayaraman, 2007).

Perhaps one of the most visible examples of suchcomplex initiatives is Walmart. In October 2005,Walmart launched an ambitious sustainability pro-gram with three broad objectives: (1) be powered by100 percent renewable energy, (2) create zerowaste, and (3) sell products that sustain people andthe environment (Walmart, 2009). The potentialbenefits of this initiative were emphasized by, forexample, the New York Times: “While the initiativemay be good for the environment, it may also begood for Wal-Mart. Driving costs out of the supplychain could result in savings for Wal-Mart that can

760 JuneAcademy of Management Journal

be passed along to consumers―enabling the com-pany to uphold its reputation as a destination forrock-bottom prices” (2010: B3). An explanation ofhow and why corporations would pursue environ-mental CSR derives from the natural resource–based view of the firm (e.g., Hart, 1995; Hart &Dowell, 2011; Russo & Fouts, 1997). This theoryrecognizes that heterogeneity of resources in a firmis a driver of competitive differences within anindustry; those companies that foster resources insupport of environmental awareness are likely togain competitive advantages and hence achievehigher profits.

In the spirit of this literature, I argue that a com-pany’s positive engagement with the environmentgenerates new and competitive resources for thefirm. Accordingly, I hypothesize a positive rela-tionship between environmental CSR and stockprices:

Hypothesis 1. Shareholders react positively tothe announcement of eco-friendly corporateinitiatives.

Conversely, a firm’s negative engagement withthe environment may decrease the firm’s competi-tive resources. In line with the previous arguments,eco-harmful activities may waste valuable re-sources (e.g., energy or material). In addition, com-panies may incur reputation losses, which in turnmay deter customers and other strategic partners.Also, legal and cleanup costs associated with eco-harmful incidents may be substantial. For example,in the case of BP’s oil spill described above, theWall Street Journal reported: “So far, it [BP] hasspent about $22 billion on cleanup and paymentsto individuals and businesses affected by the spill,and reached a civil settlement that could cost itanother $7.8 billion” (2012). For all these reasons,eco-harmful behavior may decrease the company’scompetitive resources. This leads to the followinghypothesis:

Hypothesis 2. Shareholders react negatively tothe announcement of eco-harmful corporateevents.

This view of environment-as-a-resource is theunderlying framework in my analysis. In the fol-lowing, I extend this framework by arguing that thevalue of environment-as-a-resource depends onboth external norms of environmental CSR and in-ternal levels of environmental CSR.

External Pressure

Corporations are facing external pressure to be-have responsibly toward the environment, whichin turn may affect the value of environmental CSR.Such external pressure can come from manydifferent stakeholders. It includes, for example, en-vironmental regulations, media attention to envi-ronmental CSR, and customers’ sensitivity to envi-ronmental concerns.

Over recent decades, external pressure to engagein environmental CSR has increased tremendously.In the following, I document several stylized factsthat confirm this trend.

Environmental regulations. In their analysis ofenvironmental regulations, Allen and Shonnard(2011: 71) have documented that the number offederal environmental laws and amendments hasincreased almost continuously during the past fewdecades. In particular, they report that this numberhas increased from about 70 in the early 1980s toroughly 120 in the early 2000s.

Media attention to environmental CSR. Compa-nies’ behavior towards the environment has comeunder increasing scrutiny by the media. To obtain aquantitative proxy for media attention, I searchedthrough Factiva and counted, for each year, thenumber of unique newspaper articles that refer-enced the terms “environment” and “corporate so-cial responsibility” from five of the most widelyread newspapers (New York Times, WashingtonPost, USA Today, Wall Street Journal, and Finan-cial Times). The article counts are plotted in Fig-ure 1 (solid line) for the sample years (1980–2009).As can be seen, there has been a substantial in-crease in the number of articles on environmentalCSR over the years: while the average number ofarticles was 20 in the 1980s, it was about six timeshigher (117) in the 2000s.

Environmental performance. Parallel to the in-crease in external pressure, companies seem to beimplementing more green initiatives. This trend isconsistent with anecdotal evidence (e.g., Econo-mist, 2011) and survey evidence (e.g., Accenture &UNGC, 2010). To further document this trend to-ward stronger environmental performance, I lookedat the evolution of the KLD subindex of environ-mental strengths (described below in the Data andMethodology section) in my sample. Specifically,the average number of strengths increased from0.62 in the 1990s to 1.02 in the 2000s. This 64percent increase was significant at all reasonablesignificance levels (p � .000).

2013 761Flammer

Shareholder proposals on environmental CSR.A more direct way to quantify shareholders’ con-sciousness toward the environment is to look at thenumber of shareholder proposals pertaining to theenvironment. To do so, I used data from RiskMet-rics that covered all shareholder proposals of S&P1500 companies related to either corporate gover-nance or corporate social responsibility from 1997onward. For each proposal, the data include a shortdescription (“resolution type”) that I used to deter-mine whether a given proposal pertains to environ-mental CSR (as opposed to social responsibility). Ineach calendar year, I then counted the number ofsuch proposals and expressed it as a fraction of allproposals listed in the database. This measure isplotted in Figure 1 (dashed line) and shows that thefraction of proposals has increased substantiallyduring the study period, by roughly four times(from 1997, the starting year of the RiskMetricsdatabase, to 2009, the last year of my sample).

In addition to these stylized facts, other recentdevelopments also suggest an increase in environ-mental consciousness. One example is the rise of“green consumers” (i.e., consumers who are sup-portive of environmental causes to the extent ofswitching allegiance from one product to another,even if doing so entails higher cost) and the corre-

sponding literature on “green marketing” (e.g.,Miles & Covin, 2000). Also, the fact that CSR isbeginning to include environmental responsibility(as opposed to only social responsibility) is ratherrecent and suggests an important shift in environ-mental consciousness. This shift is reflected in, forexample, the literature on social entrepreneurship,in which environmental stewardship is increas-ingly seen as an aspect of social action (e.g., Hall,Daneke, & Lenox, 2010).

As these examples illustrate, there has been ageneral trend toward higher environmental con-sciousness over the past decades. The evidencesuggesting that companies are reacting to strongerexternal pressure by implementing more green ini-tiatives is consistent with institutional theory—theview that companies do what is most legitimized intheir field and that (changes in) institutional con-ditions may lead companies to engage in environ-mental CSR (e.g., Bansal, 2005; Bansal & Roth,2000; Barnett & King, 2008; Chatterji & Toffel, 2010;Delmas & Toffel, 2004; Hoffman, 1999, 2001; Jen-nings & Zandbergen, 1995). Similarly, research instakeholder theory has argued that stakeholderpressure may lead companies to go green (e.g.,Kassinis & Vafeas, 2006; Sharma & Henriques,2005). Although these studies focus on the motiva-

FIGURE 1Evolution of Media Attention and Shareholder Proposals Related to Environmental CSR

0%

2%

4%

6%

8%

10%

12%

14%

0

20

40

60

80

100

120

140

160

180

200

1980 1985 1990 1995 2000 2005

Newspaper articles Shareholder proposals

Shareholder Proposals Related to Environment

as a Percentage ofAll Shareholder Proposals

on Governance and Social Responsibility

Number of Newspaper Articles Mentioning Both “Environment”

and “Corporate Social Responsibility”


tion for companies to go green, much less is knownabout whether external pressure affects the rela-tionship between environmental CSR and stockprices. The view of environmental CSR as a re-source helps to characterize this relationship. Ar-guably, external pressure affects the value of envi-ronmental CSR. In particular, higher externalpressure may amplify shareholders’ negative reac-tion to the announcement of eco-harmful behavior;in times of higher environmental awareness, suchannouncements are more likely to deteriorate acompany’s reputation and scare off customers, sup-pliers, strategic partners, and others. In otherwords, the more that becoming green is institution-alized as the norm and the more that eco-friendlybehavior is widespread across firms, the moreshareholders punish companies for eco-harmful be-havior. This argument leads to the followinghypothesis:

Hypothesis 3. Shareholders’ negative reactionto the announcement of eco-harmful corporateevents increases over time.

Similarly, the more companies are enacting theinstitutional norm of going green, the lower thecompetitive value of eco-friendly initiatives, andthe less shareholders reward companies for eco-friendly behavior. Thus, I propose the followinghypothesis:

Hypothesis 4. Shareholders’ positive reactionto the announcement of eco-friendly corporateinitiatives decreases over time.

Environmental Strengths and Concerns

Arguably, the value of environment-as-a-re-source not only depends on external norms of en-vironmental CSR, but also on internal levels ofenvironmental performance, as measured by, forexample, KLD scores on environmental strengthsand concerns. To see why this is the case, I use anargument in the spirit of neoclassical economictheory. The assumption of standard neoclassicalmodels is typically that decreasing marginal re-turns of production factors (e.g., capital and labor).By the same reasoning, environmental resourcesmay exhibit decreasing marginal returns as well: Ascompanies keep “investing” in green initiatives,the marginal return of an additional green initiativedecreases. Intuitively, in the early stages of, forexample, pollution reduction, it is fairly easy andinexpensive to improve environmental perfor-

mance by harvesting the “low-hanging fruit.” As acompany’s environmental footprint improves,however, it may become progressively more diffi-cult and costly to reduce pollution.

Accordingly, companies with stronger environ-mental performance (i.e., companies with a largerstock of environmental resources) may benefit rel-atively less from the introduction of an additionalgreen initiative. Conversely, companies subject tomore severe environmental concerns may benefitrelatively more from the introduction of an eco-friendly initiative. Thus, I propose the followinghypotheses:

Hypothesis 5a. Shareholders of companieswith stronger environmental performance re-act less positively to the announcement of eco-friendly initiatives.

Hypothesis 5b. Shareholders of companiessubject to more severe environmental concernsreact more positively to the announcement ofeco-friendly initiatives.

Similarly, shareholders of companies with stron-ger environmental performance may react lessnegatively to the announcement of eco-harmfulcorporate behavior. Having a larger stock of envi-ronmental resources may act as insurance to com-fort investors that the current eco-harmful event isan anomaly rather than a pattern, thus mitigatingshareholders’ negative reaction. This reasoning isin line with the insurance-based view of CSR (e.g.,Fombrun, Gardberg, & Barnett, 2000; Godfrey,2005; Peloza, 2006), according to which CSR candevelop goodwill and trust that insures the com-pany against socially irresponsible actions.1 Con-versely, shareholders of companies that have lowerstocks of environmental resources may react morenegatively to the announcement of eco-harmfulevents, since they are less insured against suchevents. These arguments lead to the followinghypotheses:

Hypothesis 6a. Shareholders of companieswith stronger environmental performance re-act less negatively to the announcement of eco-harmful behavior.

1 Another strand of literature (e.g., Baron, 2009; Baron& Diermeier, 2007) makes the opposite claim: CSR mayincrease a company’s vulnerability as it faces increasedpublic demands and scrutiny, thereby increasing the riskof not meeting the public’s expectations.

2013 763Flammer

Hypothesis 6b. Shareholders of companiessubject to more severe environmental concernsreact more negatively to the announcement ofeco-harmful events.

DATA AND METHODOLOGY

Eco-Friendly and -Harmful Corporate Events

This study examined the stock market reaction tothe announcement of corporate news related to theenvironment. For this purpose, I used Factiva, oneof the major newspaper databases, to search theWall Street Journal (WSJ) for relevant press cover-age, and obtained the stock market data from theCenter for Research in Security Prices (CRSP). Thesample period was from January 1, 1980, to Decem-ber 31, 2009. I selected this period for the availabil-ity of its data; 1980 was the first year in whichFactiva had full coverage of the WSJ, and 2009 wasthe last year of the CRSP data.

To identify WSJ articles about environment-re-lated corporate issues, three graduate student assis-tants and I performed a search in Factiva using thefollowing keywords (variations of which are indi-cated in parentheses): “pollution,” “contamina-tion” (“radiation”), “oil spill,” “hazardous waste”(“toxic waste”), “ecosystem preservation,” “recy-cling,” “emission” (“carbon”), “global warming”(“climate change”). For each keyword, we also con-sidered basic variations (e.g., for “pollution,” wealso searched for “polluted,” “pollutes”). This waseasily done in Factiva by using “wildcards” (e.g.,searching for “pollut*,” where * is the wildcardindicator). We then read each article to ensure thatit was indeed about environment-related corporatebehavior.2

A potential concern with this analysis―as withany keyword search―is that the keyword list mightbe too narrow. Nevertheless, there is no reason tobelieve that our keyword selection would intro-duce any systematic bias into the analysis. It couldonly reduce the power of our tests (since poten-

tially relevant articles might be omitted), whichwould go against finding any significant results.

The identified articles could refer to either eco-harmful or eco-friendly corporate behavior. For ex-ample, hazardous waste is generally assumed to beharmful to the environment. However, if a com-pany decides to reduce its hazardous waste, thenthis event is considered to be eco-friendly. Accord-ingly, when reading the articles, we classified themas “eco-friendly events” or “eco-harmful events.”Articles reporting both types of behavior at thesame time were excluded.3

To obtain the final data set, I applied standarddata filters. Specifically, articles were excluded inthe following cases: (1) other significant activities(e.g., leadership changes, earnings announcements)were mentioned (see McWilliams & Siegel, 1997),(2) the company of interest was not publicly tradedat a US stock market, (3) no stock market informa-tion was available during the estimation period,and (4) the article had been previously published inthe WSJ. (I will show in robustness checks that theresults are robust to using additional data filters.)These criteria left me with a sample of 273 WSJarticles on corporate news regarding environmentalissues: 117 referring to eco-friendly events and 156referring to eco-harmful events. Appendix A listsall these events as well as the corresponding key-word category.

Having compiled the list of relevant articles, Ithen extracted the company name from each articleand matched it to the corresponding companyname in CRSP. I then used firm-level identifiersfrom CRSP to link my data set to other databases(Compustat, IBES, and KLD) that are described inthe section below on regression specification.

Event Study

The event study methodology examines the stockprice reaction to news or events. The stock marketreaction is captured by the average cumulative ab-normal return (CAR) during an “event window.”CAR is a measure of how much a stock price devi-ates from its expected value during an event win-dow. The calculation of CAR is described in detailin Appendix B.

2 In collecting the data, at least two researchers pro-cessed each keyword. Interrater agreement was 99 per-cent. In almost all cases, assessment of an article’s rele-vance was straightforward. Articles were typicallyrejected because (1) the keyword was used in a differentcontext (e.g., “contamination” can also be used in a med-ical context), or (2) they did not refer to a specific com-pany, but rather to the government (e.g., in the context oflegislations), society, a particular industry, and so on.

3 Interrater agreement for categorizing the articles was96 percent. I obtained similar results throughout includ-ing only those events with 100 percent agreement.


A large body of literature in finance has usedevent studies to quantify the market reaction to theannouncement of corporate news in the WSJ. Thisliterature has examined a broad variety of issues,including the announcement of stock splits, equityissues, credit rating downgrades, and so on (e.g., forreview articles see Kothari and Warner [2007] andMacKinlay [1997]). The common practice in thisliterature is to set the event date (i.e., day 0) as theday of an article’s publication in the WSJ. Onedrawback of this practice is that the publicationdate is not necessarily the date of the actual event,as it may have happened on the previous day (be-fore the closing of the stock exchange). This prob-lem is known as “event uncertainty” in the financeliterature. The usual method of handling this issueis to expand the event window to two days (day –1and day 0), thus considering the two-day interval(�1, 0) as event window (for a discussion, seeMacKinlay [1997]). Another common approach isto consider the three-day interval (�1, 1). In thisarticle, I use the two-day event window (�1, 0) inmy main specification and show that the results arevirtually identical if (�1, 1) is used instead. I alsoexperimented with longer event windows andshow that the results are robust (albeit a bit weaker)if the windows (�1, 2) and (�1, 3) are used instead.Such extensions of the event window account forthe fact that it might take time to establish thecharacteristics of certain events (e.g., eco-harmfulincidents); therefore, the market could still receiveinformation in the few days following the events(for a similar argument, see Barnett and King [2008:1169]).4

Regression Specification

To empirically examine whether the stock mar-ket reaction to the announcement of eco-friendlyand eco-harmful corporate behavior changed overtime, I report the average CAR for each of the threedecades covered by my sample (1980–89, 1990–99, and 2000–09). To refine this analysis―and,importantly, to rule out alternative stories―I alsoused a regression-based approach. Specifically, Iestimated the following regression (separately foreco-friendly and -harmful events):

CARijst � �j � �s � � � trendt � � =Xijst � �ijst,

where i indexes firms, j indexes events, s indexesindustries, and t indexes years. Event and industryfixed effects are denoted by �j and �s respectively;CAR is the individual cumulative abnormal returnin the two-day event window (�1, 0); trend is alinear time trend (i.e., trend � 1980, 1981, . . .2009); X is a vector of control variables; and � is theerror term. Heteroscedasticity-robust standard er-rors were used. (I obtain similar results if insteadstandard errors are clustered at the industry orevent level.) The coefficient of interest is �, whichmeasures how the stock market reaction evolvedover time.

The control variables in X include size, age, prof-itability, the market-to-book ratio, and the numberof analysts following a company. These variableswere obtained from Standard & Poor’s Compustat,except the number of analysts, which was obtainedfrom Thomson Reuters’ IBES. Size is the logarithmof total assets; age is the logarithm of the number ofyears since the company was first covered in Com-pustat; profitability is the return on assets (ROA),defined as is the ratio of net income to total assets;market-to-book ratio is the ratio of the market valueof equity to the book value of equity; analysts fol-lowing is the logarithm of the number of analystsfollowing the company in IBES.

Given the sample size of 117 eco-friendly and156 eco-harmful events, including industry fixedeffects, requires a broad industry classification. Ac-cordingly, I partitioned industries at the StandardIndustrial Classification (SIC) division level.5

4 Further extending the event window may be prob-lematic. Several studies in the finance literature haveshown that using longer event windows severely reducesthe power of event study tests such as z-statistics (e.g.,Brown & Warner, 1980, 1985; Campbell, Lo, & MacKin-lay, 1997; Kothari & Warner, 2007; MacKinlay, 1997).Similarly, in the management literature, McWilliams andSiegel (1997) advocated the use of an event window thatis as short as possible, arguing that the stock price may infact fully adjust within a few minutes or hours (1997:636). A key argument in their article―similar to thepower issue emphasized in the finance literature―wasthat longer event windows would be more likely to cap-ture confounding effects, making it harder to obtain reli-able statistical inference.

5 SIC divisions are broader than two-digit SIC codes.The ten SIC divisions are as follows (the correspondingtwo-digit SIC codes are indicated in parentheses): agri-culture, forestry, and fishing (01–09); mining (10–14);construction (15–17); manufacturing (20–39); transporta-tion, communications, and public utilities (40–49);

2013 765Flammer

Lastly, I included event fixed effects, defined at the“environmental issue” level (see Appendix A). In-cluding event fixed effects mitigates concerns thatunobserved heterogeneity at the event level coulddrive the results. Importantly, such fixed effectscontrol for differences in the size of the eventsacross categories (e.g., oil spills may be more det-rimental to the environment than pollution andtherefore yield more negative CARs).6

Finally, to test Hypotheses 5a, 5b, 6a, and 6b, Iaugmented the above specification by includingfirm-level measures of environmental performanceas additional explanatory variables. The KLD data-base provides two indexes of environmental perfor-mance: “environmental strengths” and “environ-mental concerns.” The index of environmentalstrengths ranges from 0 to 7, adding one indexpoint for each of seven possible strengths. Simi-larly, the index of environmental concerns rangesfrom 0 to 7, adding one index point for each poten-tial concern. The specific strengths and concernsare described in Appendix C, based on the descrip-tion in KLD (2006). To ensure that the KLD indexeswere not affected by the events, I lagged the KLDindexes by one year. Since KLD data were availableas of 1991 and covered a subset of the companies inthe sample, merging my data set with KLD datareduced the sample size to 55 eco-friendly and 47eco-harmful events.7

Table 1 provides summary statistics (means,standard deviations, and pairwise correlations) forall variables described in this section. These statis-tics are reported separately for the 117 eco-friendlyevents (upper panel of the table) and the 156 eco-harmful events (lower panel). As can be seen, the

summary statistics are suggestive of my six hypoth-eses. In particular, I find that the mean CAR ispositive (negative) for the announcement of eco-friendly (-harmful) events, in keeping with theview that shareholders reward companies for eco-friendly initiatives and punish them for eco-harm-ful behavior (Hypotheses 1 and 2). Further, thecorrelation between CAR and the time trend is neg-ative for both eco-friendly and -harmful events,suggesting that the reward for eco-friendly initia-tives has decreased over time, while punishmentfor eco-harmful behavior has increased (Hypothe-ses 3 and 4). Finally, the correlations between CARand the KLD indexes of environmental strengthsand concerns are consistent with the argument thatenvironmental CSR is a resource with decreasingmarginal returns and insurance-like features (Hy-potheses 5a, 5b, 6a, and 6b). In the next section, Iprovide more rigorous tests of my hypotheses usingthe event-study and regression-based methodolo-gies outlined above.

RESULTS

Stock Market Reaction to Environmental Issues

My event study analysis starts with a test of Hy-pothesis 1, concerning whether shareholders reactpositively to the announcement of eco-friendly cor-porate initiatives. The results are presented in theleft-hand panel of Table 2. For each event window,I report the average CAR as a percentage (with thecorresponding z-statistics in parentheses), as wellas counts of positive and negative individual CARs(with the corresponding generalized sign z-statis-tics in parentheses). In support of Hypothesis 1, theaverage CAR in the two-day event window is 0.84percent and significant at the 1 percent level (z �3.57). In addition, a large majority of the 117 indi-vidual CARs are positive (79 positive CARs versus38 negative). All other intervals before and after thetwo-day event window yielded CARs that are smalland insignificant, which confirms that the resultsare not driven by unrelated trends around theevent dates.

The announcement of eco-harmful corporate be-havior, in contrast, leads to negative abnormal re-turns. As the right-hand panel of Table 2 shows, theaverage CAR is negative (�0.65%) and significantat the 1 percent level (z � �3.49). Furthermore, 96out of 156 individual CARs are negative. Finding anegative average CAR is supportive of Hypothesis2, namely, that shareholders react negatively to the

wholesale trade (50, 51); retail trade (52–59); finance,insurance, and real estate (60–67); services (70–88); andpublic administration (91–99).

6 Including event fixed effects controls for the size ofthe events across categories. Ideally, one would also con-trol for the size of the event within each category. How-ever, on the basis of the information provided in thenewspaper reports, it is very difficult to construct a met-ric that objectively quantifies the “size” of eco-harmful or-friendly behavior and would be comparable across thevariety of events in this sample.

7 A few of the specific strengths and concerns were notsurveyed every year from 1991–2009, which could leadto inconsistencies in the measurement of environmentalperformance over time. However, I verified that similarresults are obtained if, instead of using the full indexes, Iconstruct strength and concern indexes that consist onlyof those items surveyed in all years.


announcements of eco-harmful corporate behavior.Lastly, the CARs in the intervals before and afterthe two-day event window are again all small andinsignificant.

I have performed a series of robustness checks,which I present in Table 3, that address potential

concerns. In the following section, I briefly discusseach of them in turn.

Cross-sectional correlation. Standard eventstudy methodology assumes that the sample con-sists of independent events. Since some of theevents cluster around certain dates, this assump-

TABLE 1Descriptive Statistics and Correlationsa

Variable Mean s.d. 1 2 3 4 5 6 7 8

Eco-friendly eventsb

1. CAR (�1, 0) 0.84 1.962. Trend 1,997.45 8.41 �.373. Sizec 10.24 1.91 �.29 .524. Agec 3.59 0.44 �.20 .14 .405. Profitability 0.05 0.05 �.01 .04 .08 .076. Market-to-book 2.51 2.28 .10 .07 �.05 �.09 .587. Analysts followingc 2.02 1.41 .02 .19 .14 .19 .20 .198. KLD environmental strengths 0.95 1.03 �.24 .35 .01 .19 .04 .21 �.079. KLD environmental concerns 2.49 1.73 .16 .03 .03 .26 .02 �.19 .10 .07Eco-harmful eventsd

1. CAR (�1, 0) �0.65 2.422. Trend 1,991.73 6.06 �.213. Sizec 9.75 2.00 .23 .314. Agec 3.57 0.47 .14 .18 .545. Profitability 0.04 0.07 �.05 .24 .21 .356. Market-to-book 2.10 1.53 �.12 .29 .18 .12 .407. Analysts followingc 1.25 1.50 .06 .35 .01 .13 .12 .238. KLD environmental strengths 0.75 1.01 .19 .17 .01 .10 .02 .18 �.079. KLD environmental concerns 2.74 1.57 �.27 .06 .15 .24 .04 �.14 .11 .13

a “CAR” is cumulative abnormal return; “KLD” is Kinder, Lydenberg, Domini Research & Analytics.b n � 117 (n � 55 in eco-friendly events rows 8 and 9); all correlations larger than |.18| (|.27| in rows 8 and 9) are significant at p � .05

(two-tailed test).c Natural logarithms.d n � 156 (n � 47 in eco-harmful events rows 8 and 9); all correlations larger than |.16| (|.29| in rows 8 and 9) are significant at p �

.05 (two-tailed test).

TABLE 2CARs around the Announcement of Eco-Friendly and -Harmful Corporate Behaviora

Event Time

Eco-Friendly Events Eco-Harmful Events

CARPositive:Negative CAR

Positive:Negative

(�40, �21) 0.17 (0.12) 64:53 (1.43) �0.88 (�0.92) 73:83 (�0.41)(�20, �11) 0.32 (0.62) 60:57 (0.69) 0.75 (0.39) 72:84 (�0.57)(�10, �6) �0.45 (�1.18) 55:62 (�0.23) 0.05 (�0.12) 79:77 (0.55)(�5, �2) �0.20 (�0.77) 57:60 (0.14) �0.12 (�0.86) 75:81 (�0.09)(�1, 0) 0.84** (3.57) 79:38** (4.21) �0.65** (�3.49) 60:96* (�2.49)(1, 5) 0.16 (0.38) 59:58 (0.51) �0.15 (�0.47) 74:82 (�0.25)(6, 20) �0.49 (�1.21) 54:63 (�0.42) �0.04 (�0.26) 73:83 (�0.41)

a For eco-friendly events, n � 117; for eco-harmful events, n � 156. Event time is expressed in days; “CAR” is “cumulative abnormalreturn” and is expressed as a percentage.

* p � .05** p � .01

Two-tailed tests.

2013 767Flammer

tion may be violated. Nevertheless, in row 1 ofTable 3 I show that my results are robust if z-sta-tistics are computed using the “crude dependenceadjustment” (CDA) of Brown and Warner (1980,1985).

Precision-weighted CARs. When computing theaverage CAR, each stock is given the same weight.An alternative is to compute the precision-weighted average CAR, which gives more weight toless volatile (i.e., more precisely estimated) abnor-mal returns. As is shown in row 2 of Table 3, theresults are robust to using precision-weighted aver-age CARs.

Exxon’s oil spill. As can be seen from the eventlist in Appendix A, Exxon received a lot of presscoverage regarding its environmental footprint afterthe 1989 oil spill incident. Thus, one may be con-cerned that the sample is dominated by this onefirm. A related concern is that, although all thesearticles convey relevant new information aboutExxon’s behavior, they may be all somewhat re-lated to the original oil spill incident. To ensurethat my results were not driven by these articles, Ire-estimated the CARs, excluding all events per-taining to Exxon. As is shown in row 3 of Table 3,excluding Exxon has little impact on the results.

Alternative asset-pricing models. So far, I haveused the market model to estimate abnormal re-turns (see Appendix B). A concern is that the ab-normal returns may reflect other factors (e.g., size,book-to-market, or past performance) that arepriced during the sample period. However, I show

in rows 4 and 5 of Table 3 that I obtain very similarresults if, instead of the market model, I use thethree-factor model of Fama and French (1993) orthe four-factor model of Carhart (1997).8

Industry-adjusted CARs. To ensure that the re-sults were not driven by industry effects, I reran theevent study using industry-adjusted returns at thethree-digit SIC code level (industry-adjusted re-turns are obtained by subtracting the average returnacross all stocks on a given trading day and in agiven three-digit SIC industry). As can be seen inrow 6 of Table 3, the CARs are very similar tobefore. I have also verified that the results are ro-bust if instead I define industries at the two- andfour-digit SIC level, or if I use the 48 industries ofFama and French (1997).

Confounding events. A concern with the eventstudy methodology is that confounding events maycomplicate statistical inference (McWilliams & Sie-gel, 1997). This concern is very much minimizedhere for two reasons. First, when selecting WSJ

8 The Fama-French three-factor model includes, in ad-dition to the market factor, the size factor (“small minusbig,” or “SMB”) and the book-to-market factor (“highminus low,” or “HML”). Carhart’s four-factor model ex-tends Fama-French three-factor model by the addition ofthe momentum factor (“up minus down,” or “UMD”).Including these factors is similar to controlling for size,book-to-market, and past performance in a cross-sec-tional regression. I obtained the data on the SMB, HML,and UMD factors from Kenneth French’s website.

TABLE 3Robustnessa

Variable

Eco-Friendly Events Eco-Harmful Events

CAR CAR

1. Cross-sectional correlation 0.84** (3.81) �0.65** (�3.01)2. Precision-weighted CARs 0.65** (3.27) �0.57** (�2.82)3. Excluding Exxon’s oil spill 0.89** (3.50) �0.61* (�2.49)4. Three-factor model of Fama and French 0.91** (4.70) �0.70** (�3.47)5. Four-factor model of Carhart 0.89** (4.60) �0.69** (�3.41)6. Industry-adjusted CARs 0.82** (3.89) �0.61** (�3.73)7. Excluding confounding events 0.82** (3.83) �0.60** (�3.59)8. Alternative event window: (�1, 1) 0.83** (2.88) �0.76** (�3.83)9. Alternative event window: (�1, 2) 0.67* (2.13) �0.78** (�3.17)

10. Alternative event window: (�1, 3) 0.66† (1.93) �0.62* (�2.48)

a For eco-friendly events, n � 117 (n � 105 in row 3, n � 115 in row 7); for eco-harmful events, n � 156 (n � 120 in row 3, n � 152in row 7). “CAR” is cumulative abnormal return and is expressed as a percentage.

† p � .10* p � .05

** p � .01Two-tailed tests.


articles, I excluded articles that referenced othersignificant corporate events (e.g., leadershipchanges, earnings announcements; see the Data andMethodology section). Second, the short (two-day)event window considered in this study reduces thelikelihood of confounding events. Nevertheless,one remaining concern was that confoundingevents not mentioned in the WSJ articles might beaffecting the results. To address this concern, I useddata from CRSP on dividend announcements, fromIBES on earnings announcements, and from SDCPlatinum on acquisition and merger announce-ments. I then re-estimated the CARs excludingevents for which such announcements coincidedwith the two-day event window. As is shown inrow 7 of Table 3, the CARs are virtually identical.

Alternative event windows. In my main specifi-cation, I relied on a two-day event window, follow-ing common practice in the finance literature forconducting event studies around the publicationdate of WSJ articles (e.g., see MacKinlay, 1997).Nevertheless, in rows 8–10 of Table 3, I show thatthe results are robust if instead I use the three-daywindow (�1, 1), the four-day window (�1, 2), orthe five-day window (�1, 3).

Changes in Shareholders’ Reaction toEnvironmental Issues over Time

To study how the relationship between environ-mental CSR and stock prices has evolved over time,I repeat the analysis in Table 2 separately for eventsthat occurred in the first decade of the sample(1980–89), the second decade (1990–99), and thethird decade (2000–09).

The results are presented in Table 4. For eco-harmful events, I found that the average CAR in thetwo-day event window for the years 1980–89 was–0.42 percent, which is insignificant (z � �1.11).

The negative impact of eco-harmful behavior be-comes stronger and increasingly significant overtime. For the years 1990–99, average CAR was–0.66 percent, which is significant at the 5 percentlevel (z � �2.53). For the most recent years (2000–09), average CAR was –1.12 percent, which is sig-nificant at the 1 percent level (z � �2.69). Thispattern is supportive of Hypothesis 3, which pre-dicts that, over time, companies are increasinglypenalized by their shareholders for irresponsiblebehavior toward the environment.

As for eco-friendly events, the stock price in-crease was strongest for 1980–89; specifically, theaverage CAR was 1.19 percent, which is significantat the 1 percent level (z � 2.72). Over time, thispositive stock market reaction has monotonicallydecreased: the average CAR was 0.89 percent (z �2.04) for 1990–99 and 0.68 percent (z � 1.76) for2000–09. Interestingly, although the stock marketreaction to the announcement of eco-friendly ini-tiatives had weakened over time, the effect was stillpositive and significant for the most recent decade.This overall pattern is consistent with Hypothesis4, which predicts that the reward for eco-friendlybehavior decreases over time.9

To assess the robustness of these findings―andin particular to rule out alternative explanations ofthe results―I used the regression-based approachdescribed under methodology; in other words, Iregressed the two-day CAR on a time trend andcontrols. The results of this regression are pre-sented in Table 5 for three variations of the regres-sion specification introduced above. In model 1, Iinclude only firm-level controls. In model 2, I alsoinclude industry fixed effects. Finally, in model 3,I further include event fixed effects. Including firm-level controls mitigates concerns that the resultsare the outcome of omitted time trends in thesecharacteristics. Including industry and event fixedeffects addresses the issue of unobserved heteroge-neity at the industry and event level, respectively.

The results for eco-harmful events are reported inthe left-hand panel of Table 5. As is shown, inmodel 1 the coefficient on the time trend is –0.04percent and highly significant (t � 3.52). This co-efficient corresponds to a decrease of �0.40 percent

9 For brevity, in Table 4 I do not report CARs in thewindows preceding and following the two-day eventwindow (�1, 0). However, I have verified that the corre-sponding CARs are always small and insignificant. I havealso verified that the results in Table 4 are robust if Iconduct the robustness checks performed in Table 3.

TABLE 4Cumulative Abnormal Return (�1, 0) across Decadesa

TimePeriod

Eco-HarmfulEvents

Eco-FriendlyEvents

1980�89 �0.42 (�1.11) 1.19** (2.72)1990�99 �0.66* (�2.53) 0.89* (2.04)2000�09 �1.12** (�2.69) 0.68† (1.76)

a Eco-friendly events: n � 117; eco-harmful events: n � 156.† p � .10* p � .05


2013 769Flammer

(–0.04% � 10) from one decade to the next, whichis in the ballpark of what I found in Table 4 whencomparing decades.10 The coefficient on the timetrend is very similar in models 2 and 3.

In the right-hand panel of Table 5, I repeat thesame analysis for eco-friendly events. As is evident,the coefficient on the time trend is negative andsignificant, regardless of the model specification.The economic magnitudes are consistent with theCARs in Table 4. For example, in model 1, thecoefficient on the time trend is –0.025 percent,which corresponds to a decrease of �0.25 percent(–0.025% � 10) from one decade to the next. Thisdecrease is again in the ballpark of what I found inTable 4 when comparing decades.

Environmental Strengths and Concerns

Table 6 shows how the analysis changes whencontrolling for environmental performance, forwhich the KLD indexes on environmental strengthsand concerns in the year preceding an event serveas proxy. Since these indexes are only available fora subset of firms as of 1991, the relevant samplesize is smaller than in the previous analysis.11

The results for eco-friendly events are presentedin the right-hand panel of Table 6. In model 1, Ireplicated the regression from model 1 of Table 5with the smaller sample size. In model 2, I furtherincluded the two KLD indexes as explanatory vari-ables. This specification disentangles between ex-ternal and internal moderators (proxied by the timetrend and the two KLD indexes, respectively). As isshown, the coefficient on the time trend in model 1is similar to the full sample estimate from model 1of Table 5. It is smaller but still significant in model2. Most importantly, the coefficient on environ-mental strengths is significantly negative, while thecoefficient on environmental concerns is signifi-cantly positive. These findings are supportive ofHypotheses 5a and 5b’s predictions that sharehold-ers of companies with stronger environmentalperformance and fewer environmental concerns,respectively, react less positively to the announce-ment of eco-friendly events.

In the left-hand panel of Table 6, I repeated theanalysis for eco-harmful events. The coefficient onthe time trend in model 1 is again similar to the fullsample estimate from model 1 of Table 5. Thiscoefficient remains negative and significant withthe included KLD indexes as explanatory variablesin model 2. Importantly, the coefficient on environ-mental strengths is significantly positive, while thecoefficient on environmental concerns is signifi-cantly negative. These findings are consistent withHypotheses 6a and 6b, according to which share-holders of companies with stronger environmentalperformance and fewer environmental concerns,

10 I obtained qualitatively similar results if, instead ofusing a time trend as proxy for external pressure, I usedthe number of newspaper articles referencing environ-mental CSR or the number of shareholder proposals per-taining to the environment (from Figure 1).

11 Owing to the reduced sample size, I did not includeindustry and event fixed effects in the regressions.

TABLE 5Regression Analysis of Cumulative Abnormal Return (�1, 0)

Variable

Eco-Harmful Events Eco-Friendly Events

Model 1 Model 2 Model 3 Model 1 Model 2 Model 3

Time trend �0.040** (3.52) �0.043** (3.69) �0.049** (3.90) �0.025** (3.23) �0.021* (2.49) �0.026** (3.15)Size 0.410** (3.57) 0.403** (3.21) 0.634** (4.22) �0.069 (0.60) �0.147 (1.09) �0.224 (1.53)Age 0.125 (0.25) 0.235 (0.46) 0.416 (0.78) �0.526 (1.22) �0.481 (1.02) �0.579 (1.19)Profitability �0.067 (0.21) �0.059 (0.16) 0.326 (0.82) �0.284 (0.69) 0.048 (0.11) 0.210 (0.44)Market-to-book �0.173 (1.28) �0.212 (1.52) �0.235 (1.27) 0.124 (1.32) 0.036 (0.38) 0.057 (0.60)Analysts following 0.103 (0.77) 0.183 (1.20) 0.176 (1.07) 0.085 (0.69) 0.028 (0.20) 0.033 (0.23)Industry fixed effects No Yes Yes No Yes YesEvent-type fixed effects No No Yes No No YesObservations 156 156 156 117 117 117R2 .15 .20 .34 .18 .26 .36

* p � .05** p � .01

Two-tailed tests.


respectively, react less negatively to the announce-ment of eco-harmful behavior.

DISCUSSION AND CONCLUSION

Are shareholders sensitive to the announcementof eco-harmful corporate behavior and eco-friendlycorporate initiatives? And if so, has their percep-tion changed over time? My findings suggest thatthe answer to these questions is yes. In this study,I argue that a company’s positive engagement withthe environment generates new and competitiveresources for the firm. Extending this view of envi-ronment-as-a-resource, I further argue that thevalue of environmental CSR depends on both ex-ternal norms and internal levels of environmentalCSR. By developing this framework and empiri-cally testing its predictions, I obtain three maininsights.

First, in keeping with the view that environ-mental CSR is a resource for firms, I find that share-holders react positively to the announcement ofeco-friendly initiatives, and negatively to the an-nouncement of eco-harmful behavior.

Second, I argue that external pressure to becomegreen is setting the institutional norm of environ-mental CSR. The more that becoming green is in-stitutionalized as the norm, the more that eco-harmful behavior has a negative effect onperceptions of a firm, because firms are punishedfor not following the norm. Similarly, the more thatcompanies enact the institutional norm of going

green, the less that shareholders reward firms foreco-friendly initiatives. In support of these hypoth-eses, I show that, over time, the positive reaction tothe announcement of eco-friendly initiatives hassignificantly decreased, while the negative reactionto the announcement of eco-harmful behavior hassignificantly increased.

Third, I posit that environmental CSR is a re-source with decreasing marginal returns. Compa-nies with a larger stock of environmental resourcesmay benefit relatively less from implementing anadditional green initiative. At the same time, thesecompanies may experience a smaller loss in thecase of an eco-harmful event, as they are betterinsured against such events. In keeping with thesearguments, I find that the higher the stock of envi-ronmental CSR, the less that shareholders rewardcompanies for eco-friendly initiatives, and the lessthey punish them for eco-harmful behavior.

My findings make several contributions to theliterature. To the best of my knowledge, this studyis the first to theorize and provide empirical evi-dence on how the relationship between environ-mental CSR and stock prices has evolved over time.The comprehensive nature of my data set makesthis analysis possible, since it spans three decadesfrom 1980 to 2009. Also, the study is the first toexamine how shareholders’ reactions depend oncorporate environmental performance (as mea-sured by KLD scores on environmental strengthsand concerns). The specific findings on the insur-ance-like features of environmental CSR are related

TABLE 6Regression Analysis of Cumulative Abnormal Return (�1, 0), Controlling for Environmental Strengths and Concernsa

Variable Eco-Harmful Events Eco-Friendly Events

Model 1 Model 2 Model 1 Model 2

Time trend �0.052** (3.10) �0.038* (2.18) �0.034** (2.79) �0.023† (1.89)KLD environmental strengths 0.259† (1.75) �0.206† (1.90)KLD environmental concerns �0.286* (2.31) 0.128† (1.85)Size 0.294 (0.99) 0.518 (1.64) 0.073 (0.45) 0.069 (0.42)Age 1.093 (1.45) 1.407 (1.53) 0.277 (0.42) 0.343 (0.47)Profitability 0.552 (0.64) 1.506 (1.60) �0.110 (0.23) �0.426 (0.81)Market-to-book �0.344 (1.24) �0.384 (1.40) 0.172 (1.38) 0.210 (1.42)Analysts following �0.001 (0.00) 0.050 (0.20) 0.065 (0.33) �0.020 (0.10)Observations 47 47 55 55R2 0.29 0.40 0.24 0.30

a “KLD” is Kinder, Lydenberg, Domini Research & Analytics.† p � .10* p � .05


2013 771Flammer

to recent research in general CSR. In particular,they are in line with Godfrey, Merrill, and Hansen(2009), who showed that the negative stock marketreaction to the announcement of legal actionsagainst companies (e.g., patent infringements, qual-ity control issues, bribery) is significantly mitigatedfor firms that participate in institutional CSR activ-ities. Further related evidence for the insurance-based view is provided in Bansal and Clelland(2004), who showed that environmentally legiti-mate firms incur less unsystematic risk than illegit-imate firms.

As for the analysis of shareholders’ reactions tothe announcement of eco-harmful and -friendlyevents on average (i.e., the average CAR for all yearsstudied), my results are consistent with the find-ings of four related articles. First, Klassen andMcLaughlin (1996) examined whether the an-nouncements of environmental awards affect stockprices. Using a sample of 140 award announce-ments from 1987 to 1991, they found a positivestock market reaction, similar to my finding of apositive reaction following the announcement ofeco-friendly corporate initiatives. Second, Hamil-ton (1995) studied how the stock market reacted tothe release of data on toxic chemical releases by theEnvironmental Protection Agency (EPA) in June1989. He documented a decrease in stock prices, aresult consistent with this study’s finding that theannouncement of eco-harmful behavior triggers anegative stock market reaction. Similarly, Shaneand Spicer (1983) examined how the stock marketreacted to the release of eight studies conducted bythe Council on Economic Priorities between 1970and 1977 regarding companies’ pollution control.They documented a negative association betweenpollution and stock prices and showed that thisnegative association is mitigated for firms withhigher pollution control performance. Lastly,Gunthorpe (1997) investigated whether the detec-tion of illegal corporate activities affected stockprices, using a sample of 69 announcements (in-cluding three EPA violations) from 1988–92. Herresults showed a negative stock market reaction,results consistent with the findings of this study.However, shareholders’ reactions to illegal activi-ties (that include mainly white-collar violationssuch as corporate fraud and bribery) may not berepresentative of shareholders’ reactions to (illegaland legal) eco-harmful corporate events.

A limitation of my analysis―like any eventstudy―is that these results only address the short-run stock market reaction. A related question is

whether environmental CSR affects shareholdervalue and firm performance in the long run. Toexamine this question, one could regress long-runmeasures of firm value (e.g., Tobin’s Q) and firmperformance (e.g., return on assets, net profit mar-gin) on proxies for environmental CSR or, morebroadly, on proxies for general CSR (for reviews ofthe literature that examines the relationship be-tween CSR and accounting measures of perfor-mance, see Margolis, Elfenbein, and Walsh [2007]and Margolis and Walsh [2001, 2003]). However,CSR is likely endogenous with respect to firm valueand firm performance, which makes such analysisdifficult. Overcoming this challenge provides aninteresting avenue for future research.

Another interesting avenue for research iswhether the conceptual framework presented inthis article would apply to CSR in general. In par-ticular, how has shareholders’ perception of CSRevolved over time? Also, does CSR exhibit decreas-ing marginal returns?

Finally, my findings have potentially significantimplications for many areas of management, in-cluding strategy, innovation, intrapreneurship, andcorporate venturing. For instance, since sharehold-ers’ short-term reactions indicate that they valueeco-friendly behavior and view environmental CSRas a valuable resource, managers and boards ofdirectors may find it worthwhile to design andimplement effective environmental CSR policies topursue their long-term objectives. The same appliesto innovation and R&D programs targeted at im-proving environmental performance. Furthermore,since eco-friendly and -harmful business strategiesmatter to shareholders, research in managementscience could benefit from explicitly integratingenvironmental considerations into managerial de-cision making. Finally, and perhaps most impor-tantly from a policy perspective, legislators maybenefit from coordinating environmental regula-tions with companies; since legislators and share-holders seem to share basic views of the environ-ment, increased cooperation may prove to befruitful.

REFERENCES

Accenture & UNGC. 2010. A new era of sustainability:UN Global Compact–Accenture CEO study 2010.New York: United Nations Global Compact and Ac-centure.

Agle, B. R., Donaldson, T., Freeman, R. E., Jensen, M. C.,Mitchell, R. K., & Wood, D. J. 2008. Dialogue: To-


wards superior stakeholder theory. Business EthicsQuarterly, 18(2): 153–190.

Allen, D. T., & Shonnard, D. R. 2011. Sustainable engi-neering: Concepts, design and case studies. UpperSaddle River, NJ: Prentice Hall.

Ambec, S., Cohen, M. A., Elgie, S., & Lanoie, P. 2011. ThePorter hypothesis at 20: Can environmental regu-lation enhance innovation and competitiveness?Discussion Paper 11–01. Washington, DC: Resourcesfor the Future.

Ambec, S., & Lanoie, P. 2008. Does it pay to be green? Asystematic overview. Academy of ManagementPerspectives, 22(4): 45–62.

Bansal, P. 2005. Evolving sustainably: A longitudinalstudy of corporate sustainable development. Strate-gic Management Journal, 26: 197–218.

Bansal, P., & Clelland, I. 2004. Talking trash: Legitimacy,impression management, and unsystematic risk inthe context of the natural environment. Academy ofManagement Journal, 47: 93–103.

Bansal, P., & Roth, K. 2000. Why companies go green: Amodel of ecological responsiveness. Academy ofManagement Journal, 43: 717–736.

Barnett, M. L., & King, A. A. 2008. Good fences makegood neighbors: A longitudinal analysis of an indus-try self-regulatory institution. Academy of Manage-ment Journal, 51: 1150–1170.

Baron, D. 2009. A positive theory of moral management,social pressure, and corporate social performance.Journal of Economics and Management Strategy,18: 7–43.

Baron, D., & Diermeier, D. 2007. Strategic activism andnonmarket strategy. Journal of Economics and Man-agement Strategy, 16: 599–634.

Berchicci, L., & King, A. A. 2007. Postcards from theedge: A review of the business and environmentliterature. In J. P. Walsh & A. P. Brief (Eds.), Acad-emy of Management annals, vol. 1: 513–547. Essex,UK: Routledge.

Brown, S., & Warner, J. 1980. Measuring security priceperformance. Journal of Financial Economics, 8:205–258.

Brown, S., & Warner, J. 1985. Using daily stock returns:The case of event studies. Journal of Financial Eco-nomics, 14: 3–31.

Campbell, J., Lo, A., & MacKinlay, A. C. 1997. The econo-metrics of financial markets. Princeton, NJ: Prince-ton University Press.

Carhart, M. M. 1997. On persistence in mutual fundperformance. Journal of Finance, 52: 57–82.

Chatterji, A. K., & Toffel, M. W. 2010. How firms respondto being rated. Strategic Management Journal, 31:917–945.

Clelland, I. J., Dean, T. J., & Douglas, T. J. 2000. Steppingtowards sustainable business: An evaluation ofwaste minimization practices in U.S. manufacturing.Interfaces, 30(3): 107–124.

Delmas, M., & Toffel, M. W. 2004. Stakeholders andenvironmental management practices: An institu-tional framework. Business Strategy and the Envi-ronment, 13: 209–222.

Economist. 2011. Why firms go green. November 12.http://www.economist.com/node/21538083.

Etzion, D. 2007. Research on organizations and the nat-ural environment, 1992–present: A review. Journalof Management, 33: 637–664.

Fama, E. F., & French, K. R. 1993. Common risk factors inthe returns on stocks and bonds. Journal of Finan-cial Economics, 33: 3–56.

Fama, E. F., & French, K. R. 1997. Industry costs ofcapital. Journal of Financial Economics, 43: 153–193.

Fombrun, C. J., Gardberg, N. A., & Barnett, M. L. 2000.Opportunity platforms and safety nets: Corporate cit-izenship and reputational risk. Business and Soci-ety Review, 105: 85–106.

Freeman, R. E. 1984. Strategic management: A stake-holder approach. Boston: Pitman.

Friedman, M. 1962. Capitalism and freedom. Chicago:University of Chicago Press.

Friedman, M. 1970. The social responsibility of businessis to increase its profits. New York Times Magazine,September 13.

Godfrey, P. C. 2005. The relationship between corporatephilanthropy and shareholder value: A risk manage-ment perspective. Academy of Management Re-view, 30: 777–798.

Godfrey, P. C., Merrill, C. B., & Hansen, J. M. 2009. Therelationship between corporate social responsibilityand shareholder value: An empirical test of the riskmanagement hypothesis. Strategic ManagementJournal, 30: 425–445.

Gunthorpe, D. L. 1997. Business ethics: A quantitativeanalysis of the impact of unethical behavior by pub-licly traded corporations. Journal of Business Eth-ics, 16: 537–543.

Hall, J., Daneke, G., & Lenox, M. J. 2010. Sustainabledevelopment and entrepreneurship: Past contribu-tions and future directions. Journal of BusinessVenturing, 25: 439–448.

2013 773Flammer

Hamilton, J. 1995. Pollution as news: Media and stockmarket reactions to the TRI data. Journal of Envi-ronmental Economics and Management, 28: 98–113.

Hart, S. L. 1995. A natural resource-based view of thefirm. Academy of Management Review, 20: 986–1014.

Hart, S. L., & Dowell, G. 2011. A natural resource–basedview of the firm: Fifteen years after. Journal of Man-agement, 37: 1464–1479.

Hoffman, A. J. 1999. Institutional evolution and change:Environmentalism and the U.S. chemical industry.Academy of Management Journal, 42: 351–371.

Hoffman, A. J. 2001. Linking organizational and field-level analyses: The diffusion of corporate environ-mental practice. Organization and Environment,14: 133–156.

Jennings, P., & Zandbergen, P. 1995. Ecologically sustain-able organizations: An institutional approach. Acad-emy of Management Review, 20: 1015–1052.

Jones, T. M. 1995. Instrumental stakeholder theory: Asynthesis of ethics and economics. Academy ofManagement Review, 20: 404–437.

Kassinis, G., & Vafeas, N. 2006. Stakeholder pressuresand environmental performance. Academy of Man-agement Journal, 49: 145–159.

Klassen, R. D., & McLaughlin, C. P. 1996. The impact ofenvironmental management on firm performance.Management Science, 42: 1199–1214.

KLD. 2006. Getting started with KLD stats and KLD’sratings definitions. Boston: KLD Research & Analyt-ics.

Kothari, S. P., & Warner, J. B. 2007. Econometrics of eventstudies. In B. E. Eckbo (Ed.), Handbook of corporatefinance: Empirical corporate finance: 3–36. NorthHolland: Elsevier.

Lenox, M. J., King, A. A., & Ehrenfeld, J. R. 2000. Anassessment of design-for-environment practices inleading U.S. electronics firms. Interfaces, 30(1): 83–94.

Linton, J. D., Klassen, R., & Jayaraman, V. 2007. Sustain-able supply chains: An introduction. Journal of Op-erations Management, 25: 1075–1082.

MacKinlay, A. C. 1997. Event studies in economics andfinance. Journal of Economic Literature, 35: 13–39.

Margolis, J. D., Elfenbein, H. A., & Walsh, J. P. 2007. Doesit pay to be good? A meta-analysis and redirectionof research on the relationship between corporatesocial and financial performance. Working paper,Harvard Business School. Cambridge, MA.

Margolis, J. D., & Walsh, J. P. 2001. People and profits:The search for a link between a company’s socialand financial performance. Mahwah, NJ: Erlbaum.

Margolis, J. D., & Walsh, J. P. 2003. Misery loves compa-nies: Rethinking social initiatives by business. Ad-ministrative Science Quarterly, 48: 268–305.

McWilliams, A., & Siegel, D. 1997. Event studies in man-agement research: Theoretical and empirical issues.Academy of Management Journal, 40: 626–657.

Melnyk, S. A., Sroufe, R. P., & Calantone, R. 2003. As-sessing the impact of environmental managementsystems on corporate and perceived environmentalperformance. Journal of Operations Management,21: 329–351.

Miles, M. P., & Covin, J. G. 2000. Environmental market-ing: A source of reputational, competitive, and finan-cial advantage. Journal of Business Ethics, 23: 299–312.

New York Times. 2010. Wal-Mart unveils plan to makesupply chain greener. February 26: B3.

Orlitzky, M., Schmidt, F. L., & Rynes, S. L. 2003. Corpo-rate social and financial performance: A meta-anal-ysis. Organization Studies, 24: 403–441.

Peloza, J. 2006. Using corporate social responsibility asinsurance for financial performance. CaliforniaManagement Review, 48(2): 52–73.

Porter, M. E. 1991. America’s green strategy. ScientificAmerican, 264: 168.

Porter, M. E., & van der Linde, C. 1995a. Towards a newconception of environment-competitiveness rela-tionship. Journal of Economic Perspectives, 9: 97–118.

Porter, M. E., & van der Linde, C. 1995b. Green andcompetitive: Ending the stalemate. Harvard Busi-ness Review, 73(5): 120–134.

Rusinko, C. A. 2007. Green manufacturing: An evalua-tion of environmentally sustainable manufacturingpractices and their impact on competitive outcomes.IEEE Transactions on Engineering Management,54: 445–454.

Russo, M. V., & Fouts, P. A. 1997. A resource-basedperspective on corporate environmental perfor-mance and profitability. Academy of ManagementJournal, 40: 534–559.

Russo, M. V., & Harrison, N. S. 2005. Organizationaldesign and environmental performance: Clues fromthe electronics industry. Academy of ManagementJournal, 48: 582–593.


Shane, P. B., & Spicer, B. H. 1983. Market response toenvironmental information produced outside thefirm. Accounting Review, 58: 521–538.

Sharma, S., & Henriques, I. 2005. Stakeholder influenceson sustainability practices in the Canadian forestproducts industry. Strategic Management Journal,26: 159–180.

Sroufe, R. P. 2003. Effects of environmental managementsystems on environmental management practicesand operations. Production and Operations Man-agement, 12: 416–431.

Waage, S. 2007. Re-considering product design: A prac-tical “road-map” for integration of sustainability is-sues. Journal of Cleaner Production, 15: 638–649.

Wall Street Journal. 2012. U.S. nears BP settlements.June 28. http://online.wsj.com/article/SB10001424052702304830704577494864058303808.html.

Walmart. 2009. Walmart 2009 global sustainability re-port. Bentonville, AR: Walmart.

Caroline Flammer ([email protected]) is an assistantprofessor at the Richard Ivey School of Business at theUniversity of Western Ontario. She holds a Ph.D. ineconomics from the University of St. Gallen. Her researchinterests are corporate social responsibility, social entre-preneurship, and international business.

APPENDIX A

Events Listed by Type

TABLE A1Eco-Friendly Events

Company Name Environmental Issue Announcement Date

Inco Ltd. Emission 2/11/1980Southern Co. Contamination 4/21/1981Stauffer Chemical Co. Hazardous waste 5/26/1982Stanadyne Inc. Emission 3/1/1983Du Pont E I De Nemours & Co. Global warming 8/24/1983Monsanto Co. Recycling 9/17/1984United States Steel Corp. Recycling 5/31/1985Canadian Pacific Ltd. Pollution 9/18/1985Alcan Aluminum Ltd. Recycling 12/15/1987Scott Paper Co. Recycling 9/13/1988Dow Chemical Co. Recycling 9/27/1988Procter & Gamble Co. Recycling 10/27/1988Exxon Corp. Oil spill 3/27/1989British Petroleum PLC Oil spill 4/12/1989Church & Dwight Inc. Emission 4/14/1989Exxon Corp. Oil spill 4/20/1989Procter & Gamble Co. Recycling 4/21/1989Browning Ferris Industries Inc. Recycling 5/3/1989British Petroleum PLC Emission 8/16/1989Southern Co. Emission 8/17/1989Exxon Corp. Oil spill 10/24/1989Du Pont E I De Nemours & Co. Recycling 12/13/1989Monsanto Co. Hazardous waste 2/13/1990Eastman Kodak Co. Emission 3/30/1990Heinz H J Co. Recycling 4/9/1990Exxon Corp. Oil spill 4/12/1990Royal Dutch Petroleum Co. Emission 4/12/1990Mobil Corp. Recycling 5/16/1990Colgate Palmolive Co. Recycling 5/18/1990Unilever PLC Recycling 5/18/1990

Table continues

2013 775Flammer

TABLE A1(Continued)


Browning Ferris Industries Inc. Recycling 6/8/1990James River Corp. Va. Recycling 6/15/1990Browning Ferris Industries Inc. Recycling 6/27/1990Stone Container Corp. Recycling 7/17/1990Fuji Photo Film Ltd. Recycling 9/25/1990Acme United Corp. Recycling 12/4/1990Asarco Inc. Recycling 2/7/1991Exxon Corp. Oil spill 2/14/1991Cooper Industries Inc. Pollution 2/22/1991Coca Cola Co. Recycling 3/13/1991Weyerhaeuser Co. Emission 6/5/1991Mobil Corp. Emission 6/21/1991Du Pont E I De Nemours & Co. Emission 7/3/1991Coca Cola Co. Recycling 8/28/1991Ohio Edison Co. Emission 9/13/1991Sears Roebuck & Co. Recycling 11/14/1991American Cyanamid Co. Emission 1/29/1992Ashland Oil Inc. Emission 1/31/1992First Brands Corp. Recycling 2/27/1992New England Electric System Ecosystem preservation 8/4/1992Allied Signal Inc. Emission 3/3/1993Equitable Resources Inc. Emission 3/3/1993International Paper Co. Recycling 5/24/1993International Paper Co. Recycling 12/2/1993Texaco Inc. Pollution 2/16/1994Crown Cork & Seal Co. Inc. Recycling 3/30/1994Dell Computer Corp. Recycling 11/5/1996Cinergy Corp. Emission 9/24/1997Mobil Corp. Emission 5/24/1999B P Amoco PLC Global warming 5/4/2000Toyota Motor Corp. Emission 10/2/2000International Business Machs Co. Recycling 11/14/2000Dominion Resources Inc. Va. New Emission 11/17/2000U S X Marathon Group Hazardous waste 5/14/2001Eog Resources Inc. Global warming 1/15/2002Domtar Inc. Ecosystem preservation 4/25/2002Ford Motor Co. Del. Emission 8/20/2002Corning Inc. Emission 10/2/2002Staples Inc. Recycling 11/13/2002Exxon Mobil Corp. Oil spill 12/9/2002Ford Motor Co. Del. Emission 5/8/2003United Technologies Corp. Emission 6/27/2003Exxon Mobil Corp. Oil spill 8/25/2003K B Home Ecosystem preservation 8/25/2003Staples Inc. Ecosystem preservation 8/25/2003Intel Corp. Global warming 2/26/2004Toyota Motor Corp. Emission 7/7/2004Sony Corp. Global warming 7/28/2004General Motors Corp. Emission 10/12/2004Cinergy Corp. Emission 12/2/2004Ford Motor Co. Del. Global warming 3/31/2005JP Morgan Chase & Co. Global warming 4/25/2005Exxon Mobil Corp. Emission 10/12/2005Ford Motor Co. Del. Emission 12/20/2005Smithfield Foods Inc. Pollution 1/23/2006Weyerhaeuser Co. Emission 6/22/2006

Table continues


TABLE A1(Continued)


Ford Motor Co. Del. Emission 7/18/2006Wal-Mart Stores Inc. Recycling 8/21/2006Caterpillar Inc. Recycling 9/15/2006Ciba Specialty Chemicals Hdg. In. Contamination 10/2/2006Exxon Mobil Corp. Global warming 11/8/2006Nissan Motor Co. Ltd. Emission 12/12/2006Exxon Mobil Corp. Global warming 1/11/2007Du Pont E I De Nemours & Co. Emissions 2/6/2007JP Morgan Chase & Co. Global warming 2/27/2007Conocophillips Emission 4/11/2007Nissan Motor Co. Ltd. Emission 4/19/2007Citigroup Inc. Global warming 5/9/2007Dell Inc. Emission 6/6/2007Ford Motor Co. Del. Emission 7/10/2007Morgan Stanley Dean Witter & Co. Emission 8/15/2007Citigroup Inc. Recycling 9/5/2007Coca Cola Co. Recycling 9/6/2007Chevron Corp. New Global warming 10/18/2007Daimler A G Emission 11/13/2007General Motors Corp. Emission 1/14/2008Nokia Corp. Recycling 2/13/2008Environmental Power Corp. Emission 2/27/2008Wal-Mart Stores Inc. Emission 3/24/2008Anheuser Busch Cos. Inc. Recycling 4/22/2008Exxon Mobil Corp. Global warming 4/30/2008Wal-Mart Stores Inc. Ecosystem preservation 10/22/2008Coca Cola Co. Recycling 4/8/2009Royal Dutch Shell PLC B Emission 4/21/2009Exxon Mobil Corp. Global warming 7/15/2009Exelon Corp. Emission 10/19/2009American Electric Power Co. Inc. Emission 12/9/2009

2013 777Flammer

TABLE A2Eco-Harmful Events


Allied Products Corp. De. Hazardous waste 6/2/1980Dart Industries Inc. Hazardous waste 6/2/1980Allied Corp. Hazardous waste 8/11/1982General Electric Co. Hazardous waste 9/21/1982R S R Corp. Pollution 5/3/1983Dow Chemical Co. Hazardous waste 3/2/1984Nissan Motors Emissions 3/13/1984Du Pont E I De Nemours & Co. Contamination 3/27/1984Westinghouse Electric Corp. Hazardous waste 4/16/1984Todd Shipyards Corp. Hazardous waste 5/1/1984Ethyl Corp. Emissions 5/16/1984Aluminum Company American Hazardous waste 6/4/1984General Motors Corp. Emissions 6/22/1984General Motors Corp. Pollution 6/29/1984Diamond Shamrock Corp. Hazardous waste 9/10/1984L T V Corp. Emissions 9/19/1984Standard Oil Co. Ind. Oil spill 3/5/1985Rohm & Haas Co. Hazardous waste 5/22/1985British Petroleum PLC Recycling 8/9/1985Beatrice Company Hazardous waste 8/28/1985General Motors Corp. Emissions 5/8/1986Ashland Oil Inc. Pollution 6/26/1986Phelps Dodge Corp. Emissions 7/10/1986Kerr Mcgee Corp. Contamination 8/25/1986Union Carbide Corp. Hazardous waste 11/3/1986Xerox Corp. Hazardous waste 11/21/1986Mapco Inc. Hazardous waste 5/18/1987Atlantic Richfield Co. Hazardous waste 5/29/1987United Technologies Corp. Hazardous waste 7/22/1987International Technology Corp. Hazardous waste 9/14/1987Chrysler Corp. Emissions 12/24/1987Commercial Metals Co. Hazardous waste 6/28/1988Ashland Oil Inc. Oil spill 7/7/1988Commercial Metals Co. Hazardous waste 7/29/1988Chevron Corp. Hazardous waste 9/2/1988General Host Corp. Pollution 11/25/1988Tenneco Inc. Contamination 12/8/1988Exxon Corp. Oil spill 3/28/1989Exxon Corp. Oil spill 3/30/1989Exxon Corp. Oil spill 3/31/1989Exxon Corp. Oil spill 4/5/1989Exxon Corp. Oil spill 4/11/1989Exxon Corp. Oil spill 5/3/1989Exxon Corp. Oil spill 5/17/1989Exxon Corp. Oil spill 5/19/1989British Petroleum PLC Pollution 7/6/1989Exxon Corp. Oil spill 7/7/1989Exxon Corp. Oil spill 7/26/1989Exxon Corp. Oil spill 7/31/1989Exxon Corp. Oil spill 8/16/1989Exxon Corp. Oil spill 8/17/1989Exxon Corp. Oil spill 9/5/1989Exxon Corp. Oil spill 9/15/1989Detrex Corp. Pollution 10/24/1989Diceon Electronics Inc. Hazardous waste 11/2/1989Exxon Corp. Oil spill 1/5/1990Exxon Corp. Oil spill 1/8/1990

Table continues


TABLE A2(Continued)


British Petroleum PLC Oil spill 2/5/1990Exxon Corp. Oil spill 2/5/1990Exxon Corp. Oil spill 2/8/1990Exxon Corp. Oil spill 2/14/1990Exxon Corp. Oil spill 2/28/1990Exxon Corp. Oil spill 3/5/1990Diceon Electronics Inc. Hazardous waste 3/26/1990Bethlehem Steel Corp. Pollution 4/6/1990Chrysler Corp. Emissions 5/8/1990Coors Adolph Co. B Pollution 6/20/1990Exxon Corp. Oil spill 9/10/1990Occidental Petroleum Corp. Hazardous waste 9/13/1990Exxon Corp. Oil spill 10/1/1990Rockwell International Corp. Hazardous waste 11/28/1990Coca Cola Co. Recycling 12/5/1990Exxon Corp. Oil spill 1/31/1991Exxon Corp. Oil spill 2/8/1991Exxon Corp. Oil spill 2/19/1991Du Pont E I De Nemours & Co. Emissions 2/25/1991General Motors Corp. Emissions 3/7/1991General Electric Co. Pollution 3/15/1991Chemical Waste Mgmt Inc. Hazardous waste 3/19/1991General Motors Corp. Contamination 3/19/1991Coastal Corp. Contamination 4/19/1991Exxon Corp. Oil spill 5/3/1991United Technologies Corp. Hazardous waste 5/15/1991Wheeling Pittsburgh Corp. Pollution 5/16/1991Eljer Industries Inc. Contamination 6/4/1991Ford Motor Co. Del. Emissions 6/25/1991International Paper Co. Hazardous waste 7/5/1991Unifirst Corp. Hazardous waste 7/9/1991Dial Corp Arizona Pollution 7/10/1991United Technologies Corp. Pollution 7/24/1991Boeing Co. Hazardous waste 7/29/1991Westinghouse Electric Corp. Contamination 7/30/1991Intermet Corp. Hazardous waste 8/8/1991Merck & Co. Inc. Pollution 8/9/1991British Petroleum PLC Pollution 8/23/1991Exxon Corp. Contamination 8/23/1991Coors Adolph Co. B Pollution 10/4/1991Bristol Myers Squibb Co. Emissions 10/11/1991American Cyanamid Co. Emissions 10/23/1991Stone Container Corp. Hazardous waste 11/6/1991Ford Motor Co. Del. Emissions 11/26/1991Mobil Corp. Oil spill 1/10/1992Amoco Corp. Oil spill 1/27/1992Ford Motor Co. Del. Emissions 1/27/1992United Technologies Corp. Hazardous waste 2/5/1992Corning Inc. Emissions 3/13/1992General Motors Corp. Emissions 3/30/1992Ford Motor Co. Del. Emissions 4/10/1992Rockwell International Corp. Hazardous waste 6/2/1992Monsanto Co. Pollution 6/17/1992Kinark Corp. Hazardous waste 10/8/1992Texaco Inc. Oil spill 2/11/1993Georgia Pacific Corp. Contamination 3/10/1993Asarco Inc. Pollution 4/27/1993Chevron Corp. Oil spill 5/21/1993Allied Signal Inc. Hazardous waste 6/18/1993

Table continues

2013 779Flammer

APPENDIX B

Calculation of Cumulative Abnormal Returns

The event study methodology examines the stock pricereaction around the announcement of an event. I followcommon practice by using the publication date of thecorresponding WSJ article as the event date (day 0). Fur-thermore, I account for the possibility that the eventdocumented in the WSJ article may have happened onthe previous day by including the previous trading day(day –1) in the event window (see MacKinlay, 1997).This two-day event window is denoted by (�1, 0). To seeif there is any impact of the event on the stock pricebefore or after the two-day event window, I also consid-

ered the time intervals (�40, �21), (�20, �11), (�10,�6), (�5, �2) prior to, and the time intervals (1, 5), (6, 20)after the event window. In robustness checks, I also con-sidered the event windows (�1, 1), (�1, 2), and (�1, 3).

For each firm i, I computed the abnormal returns usingthe market model. The coefficients �i and �i of the marketmodel are estimated by ordinary least square on the basisof 200 trading days prior to the first time interval (i.e., the200 trading days used in the estimation correspond to theinterval [�240, �41]) using daily return data from CRSP.Formally, I estimated

Rit � �i � �i � Rmt � eit,

TABLE A2(Continued)


Sherwin Williams Co. Hazardous waste 7/19/1993General Electric Co. Contamination 8/16/1993C S X Corp. Pollution 9/28/1993Unocal Corp. Hazardous waste 10/8/1993Coors Adolph Co. B Emissions 2/22/1994Sun Inc. Emissions 5/27/1994Tenneco Inc. Hazardous waste 8/12/1994Exxon Corp. Oil spill 9/19/1994Cambrex Corp. Hazardous waste 10/24/1994Texas Industries Inc. Hazardous waste 9/18/1995Publicker Industries Inc. Hazardous waste 1/11/1996Weirton Steel Corp. Pollution 10/31/1996Smithfield Foods Inc. Pollution 12/18/1996Viacom Inc. A Hazardous waste 3/4/1997Buffton Corp. Pollution 7/15/1997Cinergy Corp. Emissions 5/20/1998Exxon Mobil Corp. Global warming 1/10/2000Wal-Mart Stores Inc. Hazardous waste 5/3/2000Coastal Corp. Emissions 6/5/2000Willamette Industries Inc. Emissions 7/21/2000Wal-Mart Stores Inc. Pollution 6/8/2001Exxon Mobil Corp. Hazardous waste 12/14/2001Chevrontexaco Corp. Pollution 1/9/2003Toyota Motor Corp. Emissions 3/10/2003General Motors Corp. Emissions 5/8/2003General Motors Corp. Emissions 5/19/2003Exxon Mobil Corp. Oil spill 12/9/2003Ford Motor Co. Del. Hazardous waste 3/2/2004ChevronTexaco Corp. Oil spill 8/20/2004ConocoPhillips Emissions 1/28/2005FirstEnergy Corp. Pollution 3/21/2005Exxon Mobil Corp. Global warming 3/28/2005Newmont Mining Corp. Pollution 4/4/2005Exxon Mobil Corp. Contamination 6/1/2005Wal-Mart Stores Inc. Pollution 8/16/2005Wal-Mart Stores Inc. Hazardous waste 12/21/2005Dow Chemical Co. Emissions 7/17/2006Du Pont E I De Nemours & Co. Contamination 10/2/2006Exxon Mobil Corp. Global warming 1/4/2007Dow Chemical Co. Contamination 6/4/2008


where Rit is the return on the stock of company i on dayt, �i is the intercept, �i is the systematic risk of stock i,Rmt is the daily return of the equally weighted CRSPmarket portfolio, and eit is the daily risk-adjusted resid-ual for firm i. The corresponding estimated return on thestock of firm i on day t is given by

R̂it � �i � �i � Rmt.

I then calculated the abnormal daily return (AR) ofcompany i on day t as follows:

ARit � Rit – R̂it.

Finally, I computed the cumulative abnormal returns(CARs) for each time interval by summing up the abnor-mal returns within the specific time window (e.g.,[�1, 0]).

To examine whether eco-friendly and -harmful behav-ior is perceived differently by the stock market, I dividedthe sample into eco-friendly and -harmful events. Ac-cordingly, I computed the CARs separately for both typesof events. To examine whether shareholders’ perceptionshad changed over time, I further split the sample intothree decades (1980–89, 1990–99, and 2000–09) andcomputed CARs separately for each decade.

APPENDIX C

KLD’s Environmental Strengths and Concerns

The KLD index of environmental strengths consists ofthe following strengths (see KLD, 2006):● Beneficial products and services. The company derivessubstantial revenues from innovative remediation prod-ucts, environmental services, or products that promotethe efficient use of energy, or it has developed innova-tive products with environmental benefits. (The term“environmental service” does not include services withquestionable environmental effects, such as landfills, in-cinerators, waste-to-energy plants, and deep injec-tion wells.)● Pollution prevention. The company has notably strongpollution prevention programs, including both emissionsreductions and toxic-use reduction programs.● Recycling. The company either is a substantial user ofrecycled materials as raw materials in its manufacturingprocesses, or a major factor in the recycling industry.● Clean energy. The company has taken significant mea-sures to reduce its impact on climate change and airpollution through use of renewable energy and clean

fuels or through energy efficiency. It has demonstrated acommitment to promoting climate-friendly policies andpractices outside its own operations.● Communications. The company is a signatory to theCERES Principles, publishes a notably substantive envi-ronmental report, or has notably effective internal com-munications systems in place for environmental bestpractices.● Property, plant, and equipment. The company main-tains its property, plant, and equipment with above-av-erage environmental performance for its industry.● Other strength. The company has demonstrated a su-perior commitment to management systems, voluntaryprograms, or other environmentally proactive activities.

The KLD index of environmental concerns consists ofthe following concerns (see KLD, 2006):● Hazardous waste. The company’s liabilities for hazard-ous waste sites exceed $50 million, or it has recently paidsubstantial fines or civil penalties for waste managementviolations.● Regulatory problems. The company has recently paidsubstantial fines or civil penalties for violations of air,water, or other environmental regulations, or it has apattern of regulatory controversies under the Clean AirAct, Clean Water Act, or other major environmentalregulations.● Ozone-depleting chemicals. The company is among thetop manufacturers of ozone-depleting chemicals such asHCFCs, methyl chloroform, methylene chloride, orbromines.● Substantial emissions. The company’s legal emissionsof toxic chemicals (as defined by and reported to theEPA) from individual plants into the air and water areamong the highest of the companies followed by KLD.● Agricultural chemicals. The company is a substantialproducer of agricultural chemicals (i.e., pesticides orchemical fertilizers).● Climate change. The company derives substantial rev-enues from the sale of coal or oil and its derivative fuelproducts, or indirectly from the combustion of coal or oiland its derivative fuel products. Such companies includeelectric utilities, transportation companies with fleets ofvehicles, auto and truck manufacturers, and other trans-portation equipment companies.● Other concern. The company has been involved in anenvironmental controversy that is not covered by otherKLD ratings.

2013 781Flammer

Copyright of Academy of Management Journal is the property of Academy of Managementand its content may not be copied or emailed to multiple sites or posted to a listserv withoutthe copyright holder's express written permission. However, users may print, download, oremail articles for individual use.

Documents

CORPORATE SOCIAL RESPONSIBILITY AND SHAREHOLDER …sites.bu.edu/cflammer/files/2018/09/Flammer_AMJ2013.pdf · that external pressure to becoming green (e.g., en-vironmental regulations,